Candida species are normal commensals of the oral cavity, mucosal surfaces, and gastrointestinal tract in healthy subjects but under physiological and immunological changes in the host, can cause a variety of maladies ranging from mucosal infections of the tissues and skin to life threatening systemic disease. Candida albicans is the major fungal pathogen responsible for oral candidiasis which commonly occurs in immunosuppressed patients, diabetics, patients taking antibiotics, infants, smokers and HIV-positive patients. Diagnosis of oral candidiasis is complicated, however, by the fact that current diagnostic methodologies are time consuming, may not differentiate between the different Candida species, and most importantly, do not distinguish between asymptomatic Candida carriers and patients with oral candidiasis; this can result in misdiagnosis leading to increased morbidity, increased expense and the over-prescription of antifungals. The goals of this Phase I application are to identify C. albicans gene biomarkers which are expressed solely during oral infection. Using a unique gnotobiotic mouse model of oral candidiasis, we will use suppressive subtracted hybridization and global transcription profiling to identify and compare C. albicans genes expressed during oral tissue infection to genes expressed during alimentary tract colonization. Potential disease biomarkers will be verified using candidate approaches and prioritized. The ultimate goal of this work is to develop the disease biomarkers discovered in Phase I, into a quick, sensitive and specific PCR diagnostic assay to identify patients infected with Candida species and to differentiate Candida colonizing the healthy host from patients with oral candidiasis. Since Candida species are the most important fungal pathogen of the oral cavity, the rapid and accurate diagnosis of oral candidiasis using novel disease biomarkers will significantly improve oral health.